Abstract:
A method for responding to a query is implemented on at least one computing device and includes: receiving at least one query from a user device; determining a context for the at least one query, selecting at least one deep learning network (DLN) of a plurality of DLNs to process the at least one query, where the selecting is based at least on matching the context to the at least one DLN, sending at least a representation of the at least one query and the context to the at least one DLN, receiving at least one response to the at least one query from the at least one DLN, and sending the at least one response to the user device.
Abstract:
A system and method for contextually enriching a concept database. The method includes determining, based on at least one signature of a first multimedia data element (MMDE) and signatures of a plurality of third concepts stored in the concept database, at least one matching first concept among the plurality of third concepts; generating a reduced representation of the first MMDE; comparing the reduced representation of the first MMDE to signatures representing a plurality of second MMDEs to determine a plurality of matching MMDEs among the plurality of second MMDEs; generating, based on the reduced representation of the first MMDE and the signatures representing the plurality of matching MMDEs, a second concept; and generating at least one context based on the second concept and the plurality of third concepts, wherein each context includes at least one common pattern among the second concept and at least one of the plurality of third concepts.
Abstract:
A system and method for generating channels of multimedia content element. The method includes correlating between a plurality of signatures of the multimedia content to determine a plurality of contexts; correlating the plurality of contexts to determine a plurality of assumptions for the multimedia content element; clustering the assumptions to create at least one story cluster characterizing a situation featured in the multimedia content element; and adding the multimedia content element to at least one channel, wherein the at least one channel is determined based on the at least one story cluster.
Abstract:
A system and method for sharing an image showing at least one person. The method includes generating at least one signature for the image; determining, based on the generated at least one signature, at least one personal identifier of each of the at least one person, wherein each personal identifier uniquely identifies the respective person; searching, using the determined personal identifiers, for at least one sharing identifier in at least one data source, wherein each sharing identifier indicates information for sharing the image with one of the at least one person; and sending a notification indicating the image using the at least one sharing identifier.
Abstract:
A method and apparatus for symbol-space based compression of patterns are provided. The method comprises generating an output sequence responsive of an input sequence, the input sequence being of a first length and includes a plurality of symbols, by extraction of all common patterns, wherein a common pattern includes at least two symbols and the output sequence is of a second length that is shorter than the first length; and storing in a memory the output sequence as a data layer.
Abstract:
A system and method for at least partially completing a user profile. The method includes analyzing the user profile to identify at least one missing informational element in the user profile, wherein identifying the at least one missing element further comprises determining at least one concept based on the user profile and matching the determined at least one concept to a plurality of category concepts, each concept including a collection of signatures and metadata describing the concept, wherein each category concept is associated with at least one required informational element, wherein each missing informational element is one of the at least one required informational element that is not included in the user profile; sending a query for the missing informational element; and updating at least a portion of the user profile based on a response to the query.
Abstract:
A system and method for signature-based clustering of multimedia content elements. The method includes generating at least one signature for a first multimedia content element; determining, based on the generated at least one signature, at least one tag for the first multimedia content element; searching, using the determined at least one tag, for at least one matching multimedia content element cluster in at least one data source, wherein each multimedia content element cluster includes a plurality of clustered multimedia content elements sharing a common concept; and adding the first multimedia content element to each matching multimedia content element cluster.
Abstract:
A method and system for generating a complex signature respective of a multimedia data element (MMDE) are provided. The method includes partitioning the MMDE into a plurality of different minimum size MMDEs; generating, for each of the different minimum MMDEs, at least one signature, wherein generation of each at least one signature is performed by a plurality of computational cores, each computational core having at least one configurable property characterizing the core, and wherein configuration of the at least one configurable property respective of each core results in statistical independence among the plurality of cores; and assembling at least a complex signature for the MMDE comprised of a plurality of the generated signatures.
Abstract:
A system and method for generating signatures of an input multimedia data element (MMDE) are provided. The method includes identifying low-level characteristics of the input MMDE; identifying a plurality of portions of the input MMDE, wherein each portion of the plurality of portions of the input MMDE comprises an identified low-level characteristic of the input MMDE; partitioning the input MMDE into a plurality of minimum size MMDEs, wherein each minimum size MMDE comprises a portion of the plurality of portions of the input MMDE; generating a signature for each minimum size MMDE of the plurality of minimum size MMDEs; assembling at least a complex signature comprising a plurality of signatures of the minimum size MMDEs; and storing the signatures of each of the minimum size MMDEs and the complex signature in association with the multimedia data element and the plurality of minimum size MMDEs in a storage.
Abstract:
A method and system for method for generating concept structures are disclosed. The method comprises receiving a request to create a new concept structure, wherein the request includes at least a multimedia data element (MMDE) related to the new concept structure; querying a deep-content-classification (DCC) system using the MMDE to find at least one sub-concept, wherein a sub-concept is a concept structure that partially matches the received MMDE; checking if the at least one sub-concept satisfies at least one predefined logic rule; generating one or more sub-concepts from the at least MMDE; and generating the new concept structure using one or more sub-concepts out of the at least one sub-concepts that satisfies the predefined logic rule.